This project uses machine learning algorithms to classify tweets based on sentiment. By analyzing the text of tweets, we can identify patterns that help us predict the sentiment of future tweets. With this information, we can help businesses and individuals better understand how their brand or product is perceived on social media. Whether you're a marketer looking to improve your brand reputation or an individual looking to better understand your online presence, our project can help you achieve your goals. Here I have used Natural Language Toolkit for Data Preprocessing and DecisionTreeClassifier for making predicions. I have achieved 79.78% accuracy.
1. Web Application
2. Data Preprocessing and Predictive Modeling
- Importing libraries and loading the data
- Data preprocessing using Pandas and NumPy
- Data preprocessing using NLTK
- Feature extraction using CountVectorizer
- Predictive modeling with Hyperparameter tuning
- Saving the model
- Summarzing the processes in the single cell
- Building a web application using saved models and summarized code